Women, Policy and Politics

The Construction of Policy Problems

Carol Lee Bacchi, Author

• Publisher: SAGE Publications Ltd
• Year: 1999
• Print ISBN: 9780761956754, 9780761956747
• eBook ISBN: 9781446223246
• Edition: 1
• Topic: Social Sciences
• Subtopic: Social Work
• Index: Social Sciences

This book offers a powerful new approach to policy studies. Drawing on recent perspectives from social constructionism, discourse analysis, the sociology of social problems and feminism, Carol Bacchi develops a step-by-step analytical tool for deconstructing policy problems. Her `What?s the Problem?? approach encourages students to reflect critically upon the ways in which policy problems get constructed within policy debates and policy proposals.

PART ONE

WHAT’S THE PROBLEM? AN APPROACH TO POLICY

PREAMBLE

This first section of the book is designed to introduce students to relevant literature for thinking about policy and social problems. There are indeed a number of relevant literatures, generated by different disciplinary approaches. For example, there is the literature associated with the discipline most commonly designated political science or policy studies. Within this field the student will find books which talk about ‘public policy’ or ‘social policy’. Related material will address administrative behaviour and decision making in public bureaucracies. A completely separate but directly relevant literature has been generated within sociology and is commonly characterized the sociology of social problems. This literature is large and has a long history. The interest generated by this approach to thinking about social problems resulted in the founding in 1952 of an association, the Society for the Study of Social Problems, and a journal entitled Social Problems, published by University of California Press.

When I describe the sociology of social problems as a separate literature, I do not mean that there have been no intersections with political science. In fact, there have been and it is important to reflect upon what these intersections have produced. This part of the book will deal with these as they arise. It is important to note also that both policy studies and the sociology of social problems have been affected by developments in social and political theory. Both for example reflect the shift to postpositivist and postmodern interests. Or at least these trends in intellectual thought have aroused debate within the fields, earlier in the sociology of social problems, more recently in policy studies. These debates and their significance for the approach developed in this book will be raised where they become relevant.

For simplicity’s sake, Part One has been divided into three chapters, the first two dealing with political science and its sub-discipline, policy studies, while the third chapter engages with the sociology of social problems literature. The goal in each chapter is to identify and describe the varied approaches to policy problems and the implications of each approach. The larger goal is to clarify just where a What’s the Problem (represented to be)? analysis fits in relation to these varied approaches, if and where it is similar, where it is different and how.

1

POLICY STUDIES: TRADITIONAL APPROACHES

There are a number of ways to organize and label approaches to the study of policy. Janice Dudley and Lesley Vidovich (1995: 16–18) offer a useful categorization for my purposes. They place contemporary policy studies in three categories: (1) the rational comprehensive model, (2) politically rational models of policy making and (3) public choice models of policy making. They describe the rational comprehensive view as a ‘model of decision making [which] sees policy principally as a process of problem solving. The fundamental assumption of the model is that there is a best collective decision, the public interest, that can be rationally and analytically determined if the correct neutral procedure is followed.’ In contrast, politically rational models stress ‘the political nature of the policy process [original emphasis]’. Dudley and Vidovich offer the example of Charles Lindblom (1980) who challenges the view that ‘policy is a matter of problem solving [original emphasis]’. Rather, he argues ‘that for most policy decisions there can never be any final solution, and that the problem will remain unresolved whichever policy is implemented. The policy maker can attempt only to improve the situation.’ Lindblom therefore offers a model of incrementalism in place of the ‘scientific’ rational comprehensive model. A key difference in the approaches is the attitude towards values. In the rational comprehensive model it is assumed that ‘a stable set of values and objectives is clarified and specified at the first stage of the process – referred to as defining the “decision space”’. By contrast, the political rationalist sees this as a ‘practical impossibility as negotiation and compromise between complementary and contradictory values and objectives is continuous throughout the decision making process’ (Dudley and Vidovitch, 1995: 19).

The first category then lines up with approaches generally described as the ‘rational approach to policy’. The approach embraces a positivist epistemology, endorsing the view that there is a real world which is accessible to objective description and analysis. The process of making policy is set out in clear-cut stages: formulation, implementation, evaluation. There is an assumption that there is some readily identifiable social/economic problem which needs addressing and that policy makers get together and do their best to come up with a policy which will address this problem. A further assumption is that they will approach this task rationally and come up with the best solution given cultural, political and economic constraints. The emphasis on measurement and technical expertise leads to the labelling, in some accounts, of this approach as ‘technical rationalism’. Marshall Dimock is representative of this approach. According to Dimock,

[F]irst, there are always the problems and the issues. Second, there are the facts and analyses that need to be applied to the issues. Third, there is the setting forth of alternatives and the pros and cons applicable to each possible solution – all this in the light of larger institutional goals and objectives. Fourth, there is the decision proper, which depends upon choosing among alternatives …. (1958: 140, cited in Braybrooke and Lindblom, 1963: 38)

Jan Tinberger (1956: 6–11, cited in Braybrooke and Lindblom, 1963: 38) illustrates the way in which these theorists assume that value decisions are relatively straightforward and need to precede policy analysis. He describes the opening procedures for an ideal analytical process as, first, ‘[T]he policy-maker should pursue an agreed-upon set of values’, and second, ‘the aims of policy should be clearly formulated in advance of choosing among alternative policies.’

The second category, political rationalists, are more generally described as incrementalists and pluralists. Theorists in this category object to the impression conveyed by technical rationalists that policy is a straightforward matter of finding technical answers to readily identified problems. They are much more sensitive to the give and take of politics, to the shifting of positions and perspectives, and to the role played by politics, meaning here party politics and bureaucratic politics, in decision making. Importantly, they address the need to talk about the role of values in policy making, and the most interesting in the group have tried to devise approaches which deal rationally with normative differences of opinion (Fischer, 1980; Fischer and Forester, 1987).¹ These authors insist that it is inadequate simply to see values as personal preferences which can be ignored because little can be done about them, which is one possible interpretation of some comprehensive rationalists. Rather, they insist that values can be dealt with in a rational way. Some in this group claim the label post-positivist and argue that they reject a fact-value distinction, and hence positivism. Writers in this group (Lindblom, 1980; Wildavsky, 1979) argue that a system of negotiation and compromise works well in democracies which, ideally at least, give voice to a number of groups. Their chief goal is to find ways to keep the system honest through concerted attempts to ensure that the process of decision making is as open as possible, and through measures to increase access to influence for less powerful groups. Hence, they are true pluralists. To quote Dudley and Vidovich (1995: 32) again: ‘the assumptions of the model are that the openness of the process and the wide range of participants, together with the appropriate use of technical knowledge, will result in a satisfactory policy solution or settlement.’ By way of contrast, for technical rationalists, ‘[T]he need is for expertise rather than for political participation.’

Public choice theorists are mainly concerned with improving political processes. In the main they accept that politics matters and that the task of analysts is to figure out the ways in which it matters so that less desirable aspects can be controlled for. For example, a primary concern is the way in which bureaucracies grow due to the self-interest of bureaucrats in keeping themselves employed and important. In the public choice scenario, this ‘fact’ calls for measures to curtail the growth of bureaucracy. The rational choice emphasis on rational self-interest as the determining motivation in human behaviour is obvious here (Sibeon, 1996: 28).

The limitation on the categorization produced by Dudley and Vidovich is that they do not offer a way to think about more recent developments in policy literature. They raise important themes – the relationship between policy, power, ideology, language and discursive frameworks, and indeed offer an analysis of policy making which resonates with the one offered in this book. They insist, for example, that ‘[P]olicy making is a struggle over meaning and significance.’ But they do not canvass authors who offer analyses similar to their own. Others (Hawkesworth, 1988; Pal, 1996) have attempted to group these new approaches under the heading postpositivist or postmodernist. A problem here is that you can indeed call yourself a postpositivist and still hold positivist presuppositions (see Hawkesworth, 1988: 16–17, 65). For these reasons, and given the particular focus of the approach developed in this book, I intend to categorize approaches to policy by the way in which different authors deal with policy problems. Here there will be some overlap with the categories developed by Dudley and Vidovich, and I will indicate these when they become important. But I find this organization by approach to policy problems more useful because it allows us to draw some necessary distinctions within their categories, and to clarify what is at stake in calling oneself a postpositivist.

First, I will argue that both technical rationalists and political rationalists focus on problem solution. That is, they see policy as about decision making, finding solutions to problems. Dudley and Vidovich are certainly correct that Lindblom describes his approach as opposed to a problem solving model and insists that no final solution will ever be discovered. Still, his focus is on improvement in the situation, suggesting that policy is about making decisions to improve conditions which exist in the society generally. Braybrooke and Lindblom (1963: 123) describe their approach to problem solving as ‘successive approximation’. This approach will be elaborated below.

This is not to deny the importance of the distinction between the ways technical and political rationalists approach ‘problems’. The former, as we have seen, assume that there is a discrete phase at the outset of the policy process in which values and objectives are clarified. Commonly identified as a ‘decision space’, I describe the phase as problem identification. I use this term deliberately to indicate that there is a sense in these accounts that this is a relatively unproblematic process, and that the real work, as it were, begins in finding solutions, often technical solutions, to problems which have been identified. Political rationalists, by contrast, engage in-depth in analysis of problem definition. The extensive literature here (see, for example, Dery, 1984)² makes the case that this is indeed a difficult task and tries to make suggestions about how to do it better. This group is, as mentioned above, insistent that policy makers and analysts must confront normative decisions and that any pretence that this is not the case is itself an ideological position.

I see both of these positions as working within a problem solution model. That is, even those authors who talk about the difficulty of problem definition want to offer ways of ‘doing it better’ so that ‘better solutions’ can be produced, though clearly political rationalists state emphatically that no final solution will ever be found. The focus in each case, however, is on the analyst as functionary, and on deducing what needs to be known to help them function better. This is so even with those authors who recognize that analysts do not stand outside of the normative disagreements which figure in problem definition. Here, I agree with Mary Hawkesworth (1988: 13) that both technical and political rationalists tend to argue ‘that the techniques of policy analysis can contribute to democracy by generating usable knowledge for decision-makers’. The principal source of antagonism, as she identifies it, is ‘what sort of knowledge is most useful – technical or political’.

By contrast, a What’s the Problem? approach is not interested, or at least not directly, in making analysts better at their jobs. There is a completely different rationale for this kind of analysis. Basically, the goal is to understand how policy decisions close off the space for normative debate because of the impression that indeed they are the best solution to a problem. The focus then is on the closure effected by a model of policy making as decision making, or policy making as solutions to problems or, even more specifically, of policy making as working towards making improvements in situations. In all these models, analysts are given a disinterested position, to an extent. Moreover, there is some impression that they wish to improve the situation, whatever that might mean. Despite the talk among political rationalists about values, they seem to think that it should be relatively easy to decide what it means to make an improvement in a situation (see discussion below).

Much of what some authors say when they talk about problem definition is relevant to a What’s the Problem? approach and I will illustrate this shortly. But, it is important to recognize the crucial difference in perspective and task between the approach developed by these theorists and that contained in this book. Those who talk about problem definition think it can be done better and that this will lead to better solutions. A What’s the Problem? approach, in contrast, emphasizes the inability to separate ‘solution’ from ‘problem definition’. Hence, it sets as a task, not identifying how to do problem definition better, but revealing the assumptions about the nature of the problem in any postulated solution. This focus, I argue, is more useful because it allows us to see just what is at stake in postulated solutions. It suggests that it is naive to believe that being aware of the stakes in problem definition will allow analysts to do it ‘better’. This, I would contend, is the assumption which lies behind political rationalism. Put bluntly, a What’s the Problem? approach highlights the interests and commitments at stake in postulated solutions, and suggests that analysts as well as other political actors have interests and commitments here which cannot be denied. Analysts cannot then be ‘taught’ to stand back and to ‘balance’ disagreements about values. They are in a direct sense ‘interested’.

The best way to characterize a What’s the Problem? approach, in order to clarify its relationship to the positions just described, is to say that it focuses upon problem representation. That is, it argues that every postulated ‘solution’ has built into it a particular representation of what the problem is, and it is these representations, and their implications, we need to discuss. Schematically then I would set up approaches to policy in this way:

1 Problem solution:

(a)	problem identification (comprehensive rationalists);
(b)	problem definition (political rationalists).

2 Problem representation (a What’s the Problem? approach).

Before I elaborate what is involved in a theory of problem representation and the procedures of analysis it recommends, I will look briefly at some exemplars in the categories discussed above. The purpose here is to clarify what they have to say about policy problems and how a What’s the Problem? approach differs.

No further comment will be made on public choice theorists. While this may seem odd given the popularity of the approach, public choice theorists have little to say about the nature of policy problems. Their focus rather is on the behaviour of electors or administrators and the role of self-interest in this behaviour. The more sophisticated studies mention things like ‘preference-shaping strategies’ (Dunleavy, 1991: 8), drawing attention to the ways in which political actors shape the preference of electors. The implication here is that it is too simple to talk in this case about pure self-interest. However, the range of preference-shaping strategies canvassed include only things like ‘partisan social engineering’ and ‘institutional manipulation’ (Dunleavy, 1991: 121–3). There is some discussion of agenda setting but we are left to conclude that the only motivation for any manipulation of this kind is retaining or regaining political or institutional power. A What’s the Problem? approach identifies more complex motivations behind agendas. Given the difference in focus, it is not surprising that public choice theorists see little reason to probe the meaning of problem or issue representation.

Comprehensive Rationalists

In a sense, comprehensive rationalists are not all that concerned with the nature of a problem. They are more directly concerned with describing processes for effectively moving towards an objective. Herbert Simon makes it clear that he sees a sharp and necessary distinction between fact and value, a distinction which maps onto an administrative/legislative division. He insists that, in order for administration to function, objectives have to be set and the goal of rational administration is to establish procedures which enable the individual to approach as close as possible to meeting these objectives. The term ‘rational behaviour’ in this approach ‘refers to rationality when that behaviour is evaluated in terms of the objectives of the larger organization’ (Simon, 1961: 41). Simon is prepared to admit that every decision has an ethical component, but this stands outside what it is possible to measure and hence outside of his analysis. He argues that you can examine and judge only whether the appropriate means to particular objectives have been selected. Since values stand outside the decision making process, it ‘must start with some ethical premise that is “given”’ (Simon, 1961: 50). Stated in another way, ‘the values taken as organizational objectives must be definite, so that their degree of realization in any situation can be assessed.’ Simon therefore leaves it to legislatures to set or determine values or ethical premises, which then become the ‘givens’ of administrative behaviour. This model sets up a distinction between legislature and administration which many would challenge, and a distinction between values and efficient behaviour which many would find difficult to accept. Simon goes on to deal at a very superficial level with the processes by which values, to use his terms, are set. He (1961: 52, 56) tells us simply that ‘final responsibility for determining objectives rests with a legislative body’ and that ‘principles gradually emerge to view from a given set of problems’. In this model, problems are exogenous to the process of decision making.

A later exponent of technical rationalism, Eugene Bardach, directly addresses ‘problems of problem definition in policy analysis’. In a sense, Bardach can be seen as offering the technical rationalist reply to the political rationalists who will be discussed in the next section. He (1981: 161) opens his analysis by addressing the claim that ‘definining the problem is the hardest part of the analytic task’. Rather, he insists at the very outset, ‘the hardest part of the task is find...

Table of contents

• Cover Page
• Title
• Copyright
• Contents
• Acknowledgements
• Introduction
• Part One: What’s the Problem? An Approach to Policy
• Part Two: The Problem of Women’s Inequality
• Conclusion: The Politics of Policy Studies
• Bibliography
• Index